It is difficult to purify significant amounts of membrane proteins, and the majority of those which can be purified have proven refractory to crystallization. The difficulties inherent to every step in this process have led to a paucity of membrane protein crystal structures, with a concurrent dampening effect on drug discovery and design. If such structures were available, they would accelerate the production of new medications and allow the production of more potent and more specific drugs. For proteins which can be purified in significant amounts, one innovative approach to structure determination has been to attempt the crystallization of not just the membrane protein but the membrane protein in a complex with an antibody fragment. Unfortunately, few groups have been able to copy this approach, due in part to the difficulty in acquiring antibody fragments. We believe that phage-displayed libraries of antibody fragments and engineered scaffolds will prove to be an excellent source of affinity reagents for purifying, stabilizing, and crystallizing membrane proteins. Our proposed work contains four specific aims. First, we will screen our libraries by affinity selection and isolate a panel of antibody fragments which eleven collaborators will test in crystallization trials. Second, since it is not clear whether antibodies are the best type of affinity reagent for use in membrane protein crystallization, we will develop and test alternative scaffolds based on beta propellers and fibronectin repeats. Third, we will characterize the binding of these affinity reagents to their membrane protein targets. Finally, to enable the facile use of these antibody fragments for purification and stabilization, we will engineer variants of them such that their binding to their targets is under the control of divalent cations. Improving the ability of the scientific community to purify, stabilize, characterize, and crystallize membrane proteins has the potential to be of great use in the treatment of human diseases. For example, more than half of the medications currently on the market are thought to target a single class of membrane proteins, the G-protein coupled receptors. However, the scientific community has been unable to determine the crystal structure of any human G-protein coupled receptor.